logoalt Hacker News

dyauspitrtoday at 5:42 AM7 repliesview on HN

LLMs should be trained on and directly output binary.


Replies

klodolphtoday at 5:46 AM

On the off chance that you’re serious, that would result in disastrously bad output. The difference between “jmp $+15” and “jmp $+16” is inscrutable and the LLM would not be able to pick the right one without tooling.

That tooling is a compiler. The higher level, the better chance the LLM can be steered to good output. Machine code is hopeless, don’t bother.

show 3 replies
xiaoyu2006today at 5:52 AM

It should not. Abstraction in software engineering brings intelligence. (compression correlates to intelligence)

show 3 replies
bandramitoday at 6:11 AM

Generative algorithms have been studied for decades now and while they have led to some interesting results they're a bad fit for LLMs because there's no such thing as a "plausible" binary: a small perturbation yields an unusable result.

fulafeltoday at 6:23 AM

Technically they are, just a subset. But still a practical one, they're frequently used to produce executable files.

junior44660today at 6:57 AM

[flagged]

wahnfriedentoday at 6:01 AM

[flagged]

rvztoday at 5:53 AM

I think you forgot the "/s"